wakefield_pp: compute posterior probabilities using Wakefield's approximate...

In GRealesM/RapidoPGS: A Fast and Light Package to Compute Polygenic Risk Scores

compute posterior probabilities using Wakefield's approximate Bayes Factors wakefield_pp computes posterior probabilities for a given SNP to be causal for a given SNP under the assumption of a single causal variant.

Description

This function was adapted from its namesake in cupcake package (github.com/ollyburren/cupcake/) to no longer require allele frequencies.

Usage

wakefield_pp(beta, se, pi_i = 1e-04, sd.prior = 0.2)

Arguments

beta	a vector of effect sizes (\beta) from a quantitative trait GWAS
se	vector of standard errors of effect sizes (\beta)
pi_i	a scalar representing the prior probability (DEFAULT 1 \times 10^{-4})
sd.prior	a scalar representing our prior expectation of \beta (DEFAULT 0.2). The method assumes a normal prior on the population log relative risk centred at 0 and the DEFAULT value sets the variance of this distribution to 0.04, equivalent to a 95\ is in the range of 0.66-1.5 at any causal variant.

Value

a vector of posterior probabilities.

Author(s)

Olly Burren, Chris Wallace, Guillermo Reales

GRealesM/RapidoPGS documentation built on Oct. 15, 2023, 2:43 p.m.